Using Mobile Phones to Monitor Newborn Jaundice


© 2014 University of Washington

website by Mayank Goel

BiliCam is a smartphone-based non-invasive medical device that uses the on-device camera to monitor jaundice in newborns. Newborn jaundice manifests as a yellow discoloration of skin from excess bilirubin, a chemical byproduct of recycling old blood cells. Bilicam uses an ensemble of machine learning models from data learned from nearly 100 newborn babies to detect subtle changes in skin color.  Early detection of extreme jaundice is essential to prevent brain damage or death. Current detection techniques, however, require clinical tests with blood samples or other specialized and expensive equipment. Newborns often depend on visual assessments of their skin color at home, that is known to be unreliable.

BiliCam is a completely software solution and does not require any attachment to the phone to reliably detect bilirubin levels. In addition to the phone, it will only require a disposable color calibration card. The technology is based on the analysis of newborns’ images collected in a standardized way on the phone. The color data in the image are used to estimate the bilirubin level. The initial assessments on BiliCam are very promising, with bilirubin levels estimated with the technology matching closely with the serum level (the gold standard). The system is basically a smartphone app  and requires no special expertise for use.





BiliCam: Using Mobile Phones to Monitor Newborn Jaundice [Link]

Lilian de Greef, Mayank Goel, Min Joon Seo, Eric Larson, James W. Stout, James A. Taylor, Shwetak N. Patel

The 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2014)